Theara Seng

Theara Seng

Electronics, IoT, Robotics

Bachelor Thesis: Indoor Localization and Autonomous Navigation of a Four-Wheeled Mobile Robot

Date:

🎓 Overview

This project presents the design and implementation of an autonomous mobile robot capable of performing indoor localization and navigation using the Robot Operating System (ROS).

The robot is built on a four-wheeled mecanum platform, allowing omnidirectional movement and flexible navigation in constrained environments.

📄 Full Report: Download Thesis (PDF)

🎯 Objectives

The goal of this project is to develop a robot that can:

  • 📍 Estimate its position and orientation accurately
  • 🗺️ Build a map of unknown environments
  • 🤖 Navigate autonomously from one point to another
  • 🚫 Avoid obstacles during movement

As highlighted in the thesis, the key challenge in robotics is:

“Where am I?” and “Where am I going?” :contentReference[oaicite:0]{index=0}

🧠 Key Contributions

🔹 Sensor Fusion with EKF

  • Combined:
    • Wheel encoders
    • MPU6050 IMU
    • Android IMU
  • Implemented Extended Kalman Filter (EKF) for accurate localization

✔ Improved odometry accuracy significantly

🔹 Robot Kinematics (Mecanum Wheel)

  • Developed forward and inverse kinematic models
  • Enabled omnidirectional movement

✔ Smooth and flexible robot motion

🔹 SLAM (Simultaneous Localization and Mapping)

  • Used LiDAR sensor for mapping
  • Implemented ROS package: gmapping

✔ Robot builds map while navigating

🔹 Autonomous Navigation

  • Implemented:
    • AMCL (Localization)
    • move_base (Path Planning)

✔ Robot moves autonomously from start → goal

⚙️ System Architecture

Sensors → EKF → Odometry → SLAM → Map → AMCL → Navigation → Robot

🧰 Hardware

  • 4-Wheel Mecanum Robot
  • Wheel Encoders
  • MPU6050 IMU
  • Android IMU
  • LiDAR Scanner
  • Arduino + Raspberry Pi

💻 Software

  • ROS (Robot Operating System)
  • RViz
  • Gmapping
  • AMCL
  • Move Base
  • Python / C++

🎥 Demo

👉 Watch on YouTube

📊 Results

Accurate localization using sensor fusion Real-time environment mapping Successful autonomous navigation Stable robot motion

🔮 Future Work

Add computer vision (camera-based perception) Improve localization using multi-sensor fusion Deploy in real-world applications (logistics, service robots)

🧑‍💻 Author

Theara Seng Robotics | AI | Embedded Systems